© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33675
    Prev
    Next
    INDEX
    Explanations

    code sections and labels

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is firing on code-style tokens—especially punctuation characters, numeric literals, and uppercase assembler or directive names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ﻻ
    -1.16
    ﻷ
    -1.12
     dziecko
    -1.09
    gamot
    -1.09
    ANTAGES
    -1.05
    ό
    -1.02
     miljoen
    -1.00
    leroi
    -0.96
    :^{
    -0.94
     opportunity
    -0.94
    POSITIVE LOGITS
    Таблица
    1.13
     there
    1.10
    诡异
    1.09
     portátiles
    1.09
    修为
    1.05
    durante
    1.05
     бого
    1.04
     MEXICO
    1.03
     Medica
    1.03
    校长
    1.02
    Activations Density 0.008%

    No Known Activations